Methods of reconstructing skulls

Abstract

A three-dimensional model of a reconstructed bone framework is obtained using an iterative, surface interpolating algorithm. A reference and a target three-dimensional model are provided. The reference model is non-rigidly registered to the target model based upon positional constraints, so as to produce a registered reference model. An initial reconstructed model is set as the registered reference model. A first correspondence search is iteratively conducted to identify a first set of corresponding points on the reconstructed and target models. During each iteration, the reconstructed model is incrementally and non-rigidly registered to the target model based upon the corresponding points. A second correspondence search is conducted to identify a second set of corresponding points on the reconstructed and target models. Crossings in the identified corresponding points are removed and the reconstructed model is non-rigidly registered to the target model, based upon the remaining corresponding points so as to produce a registered, fully reconstructed model. The preferred embodiment is the reconstruction of a skull using Laplacian deformation method with flip-avoiding interpolating surface (FAIS) step.

The invention claimed is: 
     
       1. A method of obtaining a three-dimensional model of a reconstructed bone framework comprising using an iterative, surface interpolating algorithm, the method comprising the steps of:
 providing a target three-dimensional model of a bone framework to be reconstructed, the target model comprising a first three-dimensional mesh model; 
 providing a reference three-dimensional model of a bone framework, the reference model comprising a second three-dimensional mesh model; 
 non-rigidly registering the reference model to the target model based upon anatomical landmarks so as to produce a registered reference model; 
 setting an initial reconstructed model as the registered reference model; 
 iteratively conducting a first correspondence search using a first correspondence search method to identify a first set of corresponding mesh vertices on the reconstructed and target models, and during each iteration non-rigidly registering the reconstructed model incrementally to the target model based upon the corresponding mesh vertices; 
 conducting a second correspondence search using a second correspondence search method to identify a second set of corresponding mesh vertices on the reconstructed and target models; and 
 removing crossings in the identified second set of corresponding mesh vertices and non-rigidly registering the reconstructed model to the target model based upon the remaining corresponding mesh vertices so as to produce a registered, fully reconstructed model, which is the model of the reconstructed bone framework, wherein the first correspondence search method and the second correspondence search method differ such that a density of the remaining corresponding mesh vertices is greater than a density of the first set of corresponding mesh vertices from each iteration. 
 
     
     
       2. A method as claimed in  claim 1 , wherein the first correspondence search comprises determining that a distance between a pair of corresponding mesh vertices in said first set is less than or equal to a first predetermined parameter, and that surface normals of the pair of corresponding mesh vertices differ by no more than 10 degrees. 
     
     
       3. A method as claimed in  claim 2 , wherein the first predetermined parameter is 0.5 mm. 
     
     
       4. A method as claimed in  claim 1 , wherein the second correspondence search comprises determining that a distance between a pair of corresponding mesh vertices in said second set is less than or equal to a second predetermined parameter, and that a mesh vertex on the target model is the nearest surface mesh vertex to a corresponding mesh vertex on the reconstructed model. 
     
     
       5. A method as claimed in  claim 4 , wherein the second predetermined parameter is 3 mm. 
     
     
       6. A method as claimed in  claim 1 , wherein iteratively conducting a first correspondence search further comprises selecting a sparse subset of the first set by applying a no-crossing condition. 
     
     
       7. A method as claimed in  claim 6 , wherein the sparse subset contains approximately fewer than 10% of vertices on the target model. 
     
     
       8. A method as claimed in  claim 1 , wherein removing crossings further comprises obtaining a dense set of corresponding mesh vertices. 
     
     
       9. A method as claimed in  claim 8 , wherein the dense set of corresponding mesh vertices contains approximately 80% of vertices on the target model. 
     
     
       10. A method as claimed in  claim 1 , wherein anatomical landmarks are obtained from manual marking of significant anatomical landmarks on the reference model and on the target model. 
     
     
       11. A method as claimed in  claim 10 , wherein the landmarks are separated to ensure no crossing. 
     
     
       12. A method as claimed in  claim 1 , wherein the target model is a model of a defective skull and the reference model is a model of a non-defective skull. 
     
     
       13. A method as claimed in  claim 1 , wherein each of the non-rigid registrations is carried out using Laplacian deformation. 
     
     
       14. A method as claimed in  claim 12 , further comprising providing the target model based upon images of the defective skull. 
     
     
       15. A method as claimed in  claim 14 , wherein an image intensity and/or a volume of interest of the images is modified. 
     
     
       16. A method as claimed in  claim 14 , further comprising editing the target model to remove non-relevant parts. 
     
     
       17. A method as claimed in  claim 14 , further comprising marking a defective part of the target model. 
     
     
       18. A method as claimed in  claim 14 , wherein at least one of the modifying, editing and marking is carried out manually via a user interface of a software package. 
     
     
       19. A method as claimed in  claim 1 , further comprising selecting the reference model from a plurality of reference models based upon a similarity between the target model and the reference model. 
     
     
       20. A method as claimed in  claim 1 , wherein the bone framework comprises a skull.